Computer-implemented system and method that allows for the rehabilitation of individuals with strabismus by leveraging commercially available VR headset technology. Eye tracking capabilities and external cameras of existing VR headsets, enables the system and method to first gather the image that the dominant, unaffected eye is focused on (FIG. 1). Then, given where the strabismus unaffected eye's gaze falls on its screen, the system and method will take the image and identify where that point lies on the strabismus affected eye's screen. This point is the ideal gaze point of the strabismus affected eye. The software solution disclosed herein will then warp the strabismus affected eye's screen so that this new gaze will line up with the unaffected eye's gaze.
Legal claims defining the scope of protection. Each claim is shown in both the original legal language and a plain English translation.
2. The computer-implemented method as defined in claim 1, wherein the homography matrix applies point-to-point conversion computation.
A computer-implemented method for image processing involves applying a homography matrix to perform point-to-point conversion computations between two images. The homography matrix is a mathematical transformation that maps points from one image plane to another, enabling tasks such as image alignment, stitching, or perspective correction. The method addresses the challenge of accurately aligning images captured from different viewpoints or under varying conditions, which is essential for applications like panoramic imaging, augmented reality, and 3D reconstruction. By using a homography matrix, the method ensures precise geometric transformations, preserving the structural relationships between corresponding points in the images. This approach is particularly useful in scenarios where images must be seamlessly combined or overlaid, such as in medical imaging, surveillance, or autonomous navigation systems. The method may also include preprocessing steps to enhance image quality or feature extraction techniques to improve the accuracy of the homography computation. The point-to-point conversion ensures that each pixel in the source image is accurately mapped to its corresponding position in the target image, minimizing distortion and maintaining visual coherence. This technique is widely applicable in computer vision systems where geometric consistency between images is critical.
3. The computer-implemented method as defined in claim 1, wherein the homography matrix is applied and recalculated as the focal point moves.
This invention relates to computer vision techniques for adjusting a homography matrix in real-time as a focal point moves. The homography matrix is a mathematical transformation used to map points between two planes, often applied in image processing to align or rectify images from different perspectives. The problem addressed is maintaining accurate image alignment when the focal point, or the point of interest in the image, changes position. Traditional methods recalculate the homography matrix only at fixed intervals or when triggered manually, leading to misalignment during dynamic movements. The method involves continuously applying and recalculating the homography matrix as the focal point moves. This ensures that the transformation remains accurate, compensating for perspective changes in real-time. The recalculation is based on updated positional data of the focal point, which may be derived from sensor inputs, user interactions, or other dynamic sources. The method may also include preprocessing steps to refine input data, such as noise reduction or feature extraction, to improve the accuracy of the homography matrix. The recalculated matrix is then applied to the image or image sequence to maintain alignment. This approach is particularly useful in applications like augmented reality, robotics, and surveillance, where precise image registration is critical during movement.
4. The computer-implemented method as defined in claim 1, wherein highlighting the focal point of the dominant eye's image includes-eliminating excessive data by cropping said image.
This invention relates to computer vision techniques for enhancing visual perception by focusing on the dominant eye's image. The problem addressed is the presence of excessive data in visual inputs, which can obscure key details and reduce processing efficiency. The solution involves a method that highlights the focal point of the dominant eye's image by selectively cropping the image to remove irrelevant or distracting data. This cropping process ensures that only the most relevant visual information is retained, improving clarity and computational efficiency. The method may also include preprocessing steps to identify the dominant eye and determine the optimal focal point for cropping. By reducing the amount of data processed, the technique enhances both the accuracy and speed of subsequent image analysis tasks, such as object recognition or tracking. The approach is particularly useful in applications requiring real-time processing, such as augmented reality, medical imaging, or autonomous systems, where minimizing unnecessary data improves performance and reduces resource consumption. The invention leverages computational techniques to dynamically adjust the cropping parameters based on the visual context, ensuring adaptability across different scenarios.
Cooperative Patent Classification codes for this invention. Click any code to explore related patents in that topic.
September 23, 2021
May 28, 2024
Browse 5M+ US patents with plain-English claim translations and AI-generated analysis.